Synthetic oligonucleotides are widely used compounds in disparate fields such as molecular biology and DNA-based diagnostics and therapeutics.
Therapeutics
In therapeutics, e.g., oligonucleotides have been used successfully to block translation in vivo of specific mRNAs thereby preventing the synthesis of proteins which are undesired or harmful to the cell/organism. This concept of oligonucleotide mediated blocking of translation is known as the “antisense” approach. Mechanistically, the hybridising oligonucleotide is thought to elicit its effect by either creating a physical block to the translation process or by recruiting cellular enzymes that specifically degrades the mRNA part of the duplex (RNAseH).
More recently, oligoribonucleotides and oligodeoxyribonucleotides and analogues thereof which combine RNAse catalytic activity with the ability to sequence specifically interact with a complementary RNA target (ribozymes) have attracted much interest as antisense probes. Thus far ribozymes have been reported to be effective in cell cultures against both viral targets and oncogenes.
To completely prevent the synthesis of a given protein by the antisense approach it is necessary to block/destroy all mRNAs that encode that particular protein and in many cases the number of these mRNA are fairly large. Typically, the mRNAs that encode a particular protein are transcribed from a single or a few genes. Hence, by targeting the gene (“antigene” approach) rather than its mRNA products it should be possible to either block production of its cognate protein more efficiently or to achieve a significant reduction in the amount of oligonucleotides necessary to elicit the desired effect. To block transcription, the oligonucleotide must be able to hybridise sequence specifically to double stranded DNA. In 1953 Watson and Crick showed that deoxyribo nucleic acid (DNA) is composed of two strands (Nature, 1953, 171, 737) which are held together in a helical configuration by hydrogen bonds formed between opposing complementary nucleobases in the two strands. The four nucleobases, commonly found in DNA are guanine (G), adenine (A), thymine (T) and cytosine (C) of which the G nucleobase pairs with C, and the A nucleobase pairs with T. In RNA the nucleobase thymine is replaced by the nucleobase uracil (U) which similarly to the T nucleobase pairs with A. The chemical groups in the nucleobases that participate in standard duplex formation constitute the Watson-Crick face. In 1959, Hoogsteen showed that the purine nucleobases (G and A) in addition to their Watson-Crick face have a Hoogsteen face that can be recognised from the outside of a duplex, and used to bind pyrimidine oligonucleotides via hydrogen bonding, thereby forming a triple helix structure. Although making the “antigene” approach conceptually feasible the practical usefulness of triple helix forming oligomers is currently limited by several factors including the requirement for homopurine sequence motifs in the target gene and a need for unphysiologically high ionic strength and low pH to stabilise the complex.
The use of oligonucleotides known as aptamers are also being actively investigated. This promising new class of therapeutic oligonucleotides are selected in vitro to specifically bind to a given target with high affinity, such as for example ligand receptors. Their binding characteristics are likely a reflection of the ability of oligonucleotides to form three dimensional structures held together by intramolecular nucleobase pairing.
Likewise, nucleosides and nucleoside analogues have proven effective in chemotherapy of numerous viral infections and cancers.
Also, various types of double-stranded RNAs have been shown to effectively inhibit the growth of several types of cancers.
Diagnostics
In molecular biology, oligonucleotides are routinely used for a variety of purposes such as for example (i) as hybridisation probes in the capture, identification and quantification of target nucleic acids (ii) as affinity probes in the purification of target nucleic acids (iii) as primers in sequencing reactions and target amplification processes such as the polymerase chain reaction (PCR) (iv) to clone and mutate nucleic acids and (vi) as building blocks in the assembly of macromolecular structures.
Diagnostics utilises many of the oligonucleotide based techniques mentioned above in particular those that lend themselves to easy automation and facilitate reproducible results with high sensitivity. The objective in this field is to use oligonucleotide based techniques as a means to, for example (i) tests humans, animals and food for the presence of pathogenic micro-organisms (ii) to test for genetic predisposition to a disease (iii) to identify inherited and acquired genetic disorders, (iv) to link biological deposits to suspects in crime trials and (v) to validate the presence of micro-organisms involved in the production of foods and beverages.
General Considerations
To be useful in the extensive range of different applications outlined above, oligonucleotides have to satisfy a large number of different requirements. In antisense therapeutics, for instance, a useful oligonucleotide must be able to penetrate the cell membrane, have good resistance to extra- and intracellular nucleases and preferably have the ability to recruit endogenous enzymes like RNAseH. In DNA-based diagnostics and molecular biology other properties are important such as, e.g., the ability of oligonucleotides to act as efficient substrates for a wide range of different enzymes evolved to act on natural nucleic acids, such as e.g. polymerases, kinases, ligases and phosphatases. The fundamental property of oligonucleotides, however, which underlies all uses is their ability to recognise and hybridise sequence specifically to complementary single stranded nucleic acids employing either Watson-Crick hydrogen bonding (A-T and G-C) or other hydrogen bonding schemes such as the Hoogsteen mode. The are two important terms affinity and specificity are commonly used to characterise the hybridisation properties of a given oligonucleotide. Affinity is a measure of the binding strength of the oligonucleotide to its complementary target sequence (expressed as the thermostability (Tm) of the duplex). Each nucleobase pair in the duplex adds to the thermostability and thus affinity increases with increasing size (No. of nucleobases) of the oligonucleotide. Specificity is a measure of the ability of the oligonucleotide to discriminate between a fully complementary and a mismatched target sequence. In other words, specificity is a measure of the loss of affinity associated with mismatched nucleobase pairs in the target. At constant oligonucleotide size the specificity increases with increasing number of mismatches between the oligonucleotide and its targets (i.e. the percentage of mismatches increases). Conversely, specificity decreases when the size of the oligonucleotide is increased at a constant number of mismatches (i.e. the percentage of mismatches decreases). Stated another way, an increase in the affinity of an oligonucleotide occurs at the expense of specificity and vice-versa.
This property of oligonucleotides creates a number of problems for their practical use. In lengthy diagnostic procedures, for instance, the oligonucleotide needs to have both high affinity to secure adequate sensitivity of the test and high specificity to avoid false positive results. Likewise, an oligonucleotide used as antisense probes needs to have both high affinity for its target mRNA to efficiently impair its translation and high specificity to avoid the unintentional blocking of the expression of other proteins. With enzymatic reactions, like, e.g., PCR amplification, the affinity of the oligonucleotide primer must be high enough for the primer/target duplex to be stable in the temperature range where the enzymes exhibits activity, and specificity needs to be high enough to ensure that only the correct target sequence is amplified.
Given the shortcomings of natural oligonucleotides, new approaches for enhancing specificity and affinity would be highly useful for DNA-based therapeutics, diagnostics and for molecular biology techniques in general.
Conformationally Restricted Nucleosides
It is known that oligonucleotides undergo a conformational transition in the course of hybridising to a target sequence, from the relatively random coil structure of the single stranded state to the ordered structure of the duplex state.
A number of conformationally restricted oligonucleotides including bicyclic and tricyclic nucleoside analogues (FIGS. 1A and 1B in which B=nucleobase) have been synthesised, incorporated into oligonucleotide and oligonucleotide analogues and tested for their hybridisation and other properties.
Bicyclo[3.3.0] nucleosides (bcDNA) with an additional C-3′,C-5′-ethano-bridge (A and B) have been synthesised with all five nucleobases (G, A, T, C and U) whereas (C) has been synthesised only with T and A nucleobases (M. Tarköy, M. Bolli, B. Schweizer and C. Leumann, Helv. Chim. Acta, 1993, 76, 481; Tarköy and C. Leumann, Angew. Chem., Int. Ed. Engl., 1993, 32, 1432; M. Egli, P. Lubini, M. Dobler and C. Leumann, J. Am. Chem. Soc., 1993, 115, 5855; M. Tarköy, M. Bolli and C. Leumann, Helv. Chim. Acta, 1994, 77, 716; M. Bolli and C. Leumann, Angew. Chem., Int. Ed. Engl., 1995, 34, 694; M. Bolli, P. Lubini and C. Leumann, Helv. Chim. Acta, 1995, 78, 2077; J. C. Litten, C. Epple and C. Leumann, Bioorg. Med. Chem. Lett., 1995, 5, 1231; J. C. Litten and C. Leumann, Helv. Chim. Acta, 1996, 79, 1129; M. Bolli, J. C. Litten, R. Schultz and C. Leumann, Chem. Biol., 1996, 3, 197; M. Bolli, H. U. Trafelet and C. Leumann, Nucleic Acids Res., 1996, 24, 4660). DNA oligonucleotides containing a few, or being entirely composed, of these analogues are in most cases able to form Watson-Crick bonded duplexes with complementary DNA and RNA oligonucleotides. The thermostability of the resulting duplexes, however, is either distinctly lower (C), moderately lower (A) or comparable to (B) the stability of the natural DNA and RNA counterparts. All bcDNA oligomers exhibited a pronounced increase in sensitivity to the ionic strength of the hybridisation media compared to the natural counterparts. The α-bicyclo-DNA (B) is more stable towards the 3′-exonuclease snake venom phosphordiesterase than the β-bicyclo-DNA (A) which is only moderately more stable than unmodified oligonucleotides.
Bicarbocyclo[3.1.0]nucleosides with an additional C-1′,C-6′- or C-6′,C-4′-methano-bridge on a cyclopentane ring (D and E, respectively) have been synthesised with all five nucleobases (T, A, G, C and U). Only the T-analogues, however, have been incorporated into oligomers. Incorporation of one or ten monomers D in a mixed poly-pyrimidine DNA oligonucleotide resulted in a substantial decrease in the affinity towards both DNA and RNA oligonucleotides compared to the unmodified reference oligonucleotide. The decrease was more pronounced with ssDNA than with ssRNA. Incorporation of one monomer E in two different poly-pyrimidine DNA oligonucleotides induced modest increases in Tm's of 0.8° C. and 2.1° C. for duplexes towards ssRNA compared with unmodified reference duplexes. When ten T-analogues were incorporated into a 15mer oligonucleotide containing exclusively phosphorothioate internucleoside linkages, the Tm against the complementary RNA oligonucleotide was increased approximately 1.3° C. per modification compared to the same unmodified phosphorothioate sequence. Contrary to the control sequence the oligonucleotide containing the bicyclic nucleoside E failed to mediate RNAseH cleavage. The hybridisation properties of oligonucleotides containing the G, A, C and U-analogues of E have not been reported. Also, the chemistry of this analogue does not lend itself to further intensive investigations on completely modified oligonucleotides (K.-H. Altmann, R. Kesselring, E. Francotte and G. Rihs, Tetrahedron Lett., 1994, 35, 2331; K.-H. Altmann, R. Imwinkelried, R. Kesselring and G. Rihs, Tetrahedron Lett., 1994, 35, 7625; V. E. Marquez, M. A. Siddiqui, A. Ezzitouni, P. Russ, J. Wang, R. W. Wagner and M. D. Matteucci, J. Med. Chem., 1996, 39, 3739; A. Ezzitouni and V. E. Marquez, J. Chem. Soc., Perkin Trans. 1, 1997, 1073).
A bicyclo[3.3.0]nucleoside containing an additional C-2′,C-3′-dioxalane ring has been synthesised as a dimer with an unmodified nucleoside where the additional ring is part of the internucleoside linkage replacing a natural phosphordiester linkage (F). This analogue was only synthesised as either thymine-thymine or thymine-5-methylcytosine blocks. A 15-mer polypyrimidine sequence containing seven of these dimeric blocks and having alternating phosphordiester- and riboacetal-linkages, exhibited a substantially decreased Tm against complementary ssRNA compared to a control sequence with exclusively natural phosphordiester internucleoside linkages (R. J. Jones, S. Swaminathan, J. F. Millagan, S. Wadwani, B. S. Froehler and M. Matteucci, J. Am. Chem. Soc., 1993, 115, 9816).
The two dimers (G and H) with additional C-2′,C-3′-dioxane rings forming bicyclic[4.3.0]-systems in acetal-type internucleoside linkages have been synthesised as T-T dimers and incorporated once in the middle of 12mer polypyrimidine oligonucleotides. Oligonucleotides containing either G or H both formed significantly less stable duplexes with complementary ssRNA and ssDNA compared with the unmodified control oligonucleotide (J. Wang and M. D. Matteucci, Bioorg. Med. Chem. Lett., 1997, 7, 229).
Dimers containing a bicyclo[3.1.0]nucleoside with a C-2′,C-3′-methano bridge as part of amide- and sulfonamide-type (I and J) internucleoside linkages have been synthesised and incorporated into oligonucleotides. Oligonucleotides containing one or more of these analogues showed a significant reduction in Tm compared to unmodified natural oligonucleotide references (C. G. Yannopoulus, W. Q. Zhou, P. Nower, D. Peoch, Y. S. Sanghvi and G. Just, Synlett, 1997, 378).
A trimer with formacetal internucleoside linkages and a bicyclo[3.3.0]glucose-derived nucleoside analogue in the middle (K) has been synthesised and connected to the 3′-end of an oligonucleotide. The Tm against complementary ssRNA was decreased by 4° C., compared to a control sequence, and by 1.5° C. compared to a sequence containing two 2′,5′-formacetal linkages in the 3′-end (C. G. Yannopoulus, W. Q. Zhou, P. Nower, D. Peoch, Y. S. Sanghvi and G. Just, Synlett, 1997, 378).
Very recently oligomers composed of tricyclic nucleoside-analogues (L) have been reported to show increased duplex stability compared to natural DNA (R. Steffens and C. Leumann (Poster SB-B4), Chimia, 1997, 51, 436).
An attempt to make the bicyclic uridine nucleoside analogue Q planned to contain an additional O-2′,C-4′-five-membered ring, starting from 4′-C-hydroxymethyl nucleoside P, failed (K. D. Nielsen, Specialerapport (Odense University, Denmark), 1995).
Until now the pursuit of conformationally restricted nucleosides useful in the formation of synthetic oligonucleotides with significantly improved hybridisation characteristics has met with little success. In the majority of cases, oligonucleotides containing these analogues form less stable duplexes with complementary nucleic acids compared to the unmodified oligonucleotides. In other cases, where moderate improvement in duplex stability is observed, this relates only to either a DNA or an RNA target, or it relates to fully but not partly modified oligonucleotides or vice versa. An appraisal of most of the reported analogues are further complicated by the lack of data on analogues with G, A and C nucleobases and lack of data indicating the specificity and mode of hybridisation. In many cases, synthesis of the reported monomer analogues is very complex while in other cases the synthesis of fully modified oligonucleotides is incompatible with the widely used phosphoramidite chemistry standard.